← Models

Model profile

Deepseek R1 0528 Qwen3 8B

2025-05-29release date
2eval lineages

Evidence summary

Published evidence spans 2 evals and 1 of 7 behavior components. Its strongest relative result is AA-Omniscience (hallucination_rate, #208 of 311); its weakest is BrokenMath (sycophancy, #8 of 9).

Compare this model

Only models sharing at least one published sub-eval are listed.

Official and reference links

Published eval results

Rank is within that sub-eval. Black marks the observed result; the grey dot marks the value implied by the global rank. Values stay on each source’s native scale.

Eval / sub-evalRankValueDistributionBetterSource
AA-Omnisciencehallucination_rate#208 / 3110.865↓ lowerSource ↗official
BrokenMathsycophancy#8 / 956.3↓ lowerSource ↗official